Volumetric bioluminescence imaging of cellular dynamics with deep learning based light-field reconstruction

Abstract

The application of genetically encoded fluorophores for microscopy has afforded one of the biggest revolutions in the biosciences. Bioluminescence microscopy is an appealing alternative to fluorescence microscopy, because it does not depend on external illumination, and consequently does neither produce spurious background autofluorescence, nor perturb intrinsically photosensitive processes in living cells and animals. The low quantum yield of known luciferases, however, limit the acquisition of high signal-noise images of fast biological dynamics. To increase the versatility of bioluminescence microscopy, we present an improved low-light microscope in combination with deep learning methods to increase the signal to noise ratio in extremely photon-starved samples at millisecond exposures for timelapse and volumetric imaging. We apply our method to image subcellular dynamics in mouse embryonic stem cells, the epithelial morphology during zebrafish development, and DAF-16 FoxO transcription factor shuttling from the cytoplasm to the nucleus under external stress. Finally, we concatenate neural networks for denoising and light-field deconvolution to resolve intracellular calcium dynamics in three dimensions of freely moving Caenorhabditis elegans with millisecond exposure times. This technology is cost-effective and has the potential to replace standard optical microscopy where external illumination is prohibitive.

Deep earning enhanced bioluminescence microscopy

“…For future work, these techniques can be useful for further analysis in more complex systems such as glutamatergic neurons, zebrafish epithelial cell organization, volumetric calcium dynamics in the body wall muscle and ensemble of neurons in freely moving animals. 15 Also, spatiotemporal resolution can be improved by training with real experimental data for the suppression of errors in the images due to misalignment in the setup or optical aberrations. Additionally, preprocessing steps might allow to improve the input quality to boost the 3D reconstruction performance.…”

“…Therefore, in order to acquire high SNR data, this approach commonly demands long exposure times in the order of seconds or even tens of seconds, which limits the applications to slow biological dynamics and gene expression analyses. 11 Convolutional neural networks have been widely applied in microscopy research and have shown to outperform classical algorithms in different tasks such as segmentation, 12 denoising, [13][14][15] 3D reconstruction methods, 15,16 and superresolution. 17 The main reason for this is that neural networks take advantage of the knowledge available about the data.…”